Personalised assistance for fuel-efficient driving

Ekaterina Gilman, Anja Keskinarkaus, Satu Tamminen, Susanna Pirttikangas, Juha Röning, Jukka Riekki
2015 Transportation Research Part C: Emerging Technologies  
Recent advances in technology are changing the way how everyday activities are performed. Technologies in the traffic domain provide diverse instruments of gathering and analysing data for more fuel-efficient, safe, and convenient travelling for both drivers and passengers. In this article, we propose a reference architecture for a context-aware driving assistant system. Moreover, we exemplify this architecture with a real prototype of a driving assistance system called Driving coach. This
more » ... type collects, fuses and analyses diverse information, like digital map, weather, traffic situation, as well as vehicle information to provide drivers in-depth information regarding their previous trip along with personalised hints to improve their fuel-efficient driving in the future. The Driving coach system monitors its own performance, as well as driver feedback to correct itself to serve the driver more appropriately. j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / t r c planning a trip in such a way that the number of stops, high speeding, etc. are minimised. Ericsson et al. (2006) have calculated that a fuel-efficient route can save about 8% of fuel. Also, their study demonstrated that a fuel-efficient route is about the same as the shortest one. No significant fuel reduction effect was found for the fastest route option. Aggressive driving is another factor affecting fuel-consumption (Sivak and Schoettle, 2012; Brundell-Freij and Ericsson, 2005) . Aggressive driving means certain actions increasing the risk of road accident, like excessive speeding and improper turning. In fact, aggressive driving is found to be one of the main causes of car accidents (The AAA Foundation for Traffic Safety, 2009). Based on the related work, we may conclude that it is possible to minimise fuel consumption by discovering the most relevant factors and informing drivers how to improve their driving behaviour with respect to these factors. On the other hand, drivers vary a lot. They have different driving experience, preferences, and habits. Hence they require tailored solutions to explain what can be improved in their driving style and how (Gonder et al., 2011) . Moreover, different external factors, like traffic fluency situation, road quality, and weather may affect performance of drivers. Hence, the overall situation should be assessed to promote more fuel-efficient driving. This information is referred as context, and systems able to capture the context and react on its changes are called context-aware (Dey, 2001) . In this article, we argue that contextaware driving assistant systems provide more adequate feedback to drivers regarding fuel-efficient driving. In this article, we propose a reference architecture for context-aware driving assistance systems. This architecture is aligned with a Meta-level control framework presented by Gilman and Riekki (2012) . Their framework emphasises the necessity for self-introspective functionality for personalised and adaptive systems. This framework adds a controlling and monitoring layer to such systems. Moreover, it emphasises monitoring the overall interaction to gather feedback about how well the system supports its users in their tasks. For instance, with this kind of functionality, the system would notice that a driver constantly ignores certain advice and would perform actions to resolve such cases. The proposed architecture is exemplified with a driving support system called Driving coach. This system teaches a driver to drive better. Better driving in this context means: (1) avoiding aggressive driving, (2) trip planning, and (3) driving in a fuel-efficient manner. The system is based on real-time information, obtained from on-board sensors and external services. The driver gets feedback about his driving after each trip: comments and recommendations what to do differently in order to drive better. The key characteristics of our system are: 1. Fusion of on-board information and real-time information from third party services. 2. Identification of personal driving factors affecting the fuel use in certain situations. 3. Adaptation of the system's decision-making with respect to a driver's progress and responses to recommendations.
doi:10.1016/j.trc.2015.02.007 fatcat:5qkve33irvh6fneftqey3cjbay